MI-Fuse：面向闭源大型音频语言模型的无监督域适应的标签融合
MI-Fuse: Label Fusion for Unsupervised Domain Adaptation with Closed-Source Large-Audio Language Model
AI总结：
针对无源数据且仅API可用的LALM场景，提出MI-Fuse去噪标签融合框架，结合互信息加权与EMA教师，使学生模型在目标域超越LALM，并在六种跨域迁移中比最强基线高3.9%。
摘要：
大型音频语言模型（LALMs）在语音任务上展现出强大的零样本能力，为语音情感识别（SER）带来了希望。然而，在现实部署中，SER在域不匹配情况下常常失败，此时源数据不可用，而强大的LALM仅能通过API访问。我们提出一个问题：仅给定未标记的目标域音频和仅API可用的LALM，能否使一个学生模型适应并在目标域中超越LALM？为此，我们提出了MI-Fuse，一种去噪标签融合框架，该框架将LALM与一个源域训练的SER分类器作为辅助教师相结合。该框架从两位教师中抽取多个随机预测，通过基于互信息的的不确定性对其均值分布进行加权，并使用指数移动平均教师来稳定训练。在三个公共情感数据集和六种跨域迁移上的实验显示出一致的性能提升，学生模型超越了LALM，并比最强基线高出3.9%。该方法在不共享源数据的情况下增强了情感感知语音系统，实现了现实可行的适应。
英文摘要：
Large audio-language models (LALMs) show strong zero-shot ability on speech tasks, suggesting promise for speech emotion recognition (SER). However, SER in real-world deployments often fails under domain mismatch, where source data are unavailable and powerful LALMs are accessible only through an API. We ask: given only unlabeled target-domain audio and an API-only LALM, can a student model be adapted to outperform the LALM in the target domain? To this end, we propose MI-Fuse, a denoised label fusion framework that supplements the LALM with a source-domain trained SER classifier as an auxiliary teacher. The framework draws multiple stochastic predictions from both teachers, weights their mean distributions by mutual-information-based uncertainty, and stabilizes training with an exponential moving average teacher. Experiments across three public emotion datasets and six cross-domain transfers show consistent gains, with the student surpassing the LALM and outperforming the strongest baseline by 3.9%. This approach strengthens emotion-aware speech systems without sharing source data, enabling realistic adaptation.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
